Thursday, January 08, 2015

Japan's fertility rate is rising, and you all missed it

Today's Bat Boy Award for crazy blogging goes to Ana Swanson of the Washington Post!

The headline cries: "Japan's birth rate problem is way worse than anyone imagined". The article cites a paper by some Waseda economists, which is all about inflation and deflation, and really has nothing to do with the ostensible point of the WaPo piece. The citation of the paper is really just an excuse to show the following graph:

This is supposed to be one of those charts that shows how the government keeps getting its forecasts wrong. Swanson writes:
The data above...shows just how bad Japan has been at forecasting its fertility rate since 1965. Government projections have been almost comically wrong, as the government repeatedly interpreted the sharp decrease in the fertility rate as a temporary dip rather than a sustained trend.
Except that is not what the graph shows at all. Yes, government forecasts are way off (forecasting is hard). But for over a decade, the Japanese government has been too pessimistic about the fertility rate.

Just Look. At. The. Graph! Since 2005, the black line - the actual fertility rate - has been going up! It is now higher than the blue line representing the forecast from 2002. It is even higher than the blue like representing the forecast for 2008. The graph only looks like it goes through 2010, but a quick Google shows that the fertility rate is still over 1.4, i.e. higher than the 2002, 2006, or 2012 forecasts.

In other words, Japanese fertility has been surprising on the upside for ten years and counting!

People have been tweeting this graph, and the incorrect conclusion, all day. Didn't anyone even bother to take a look at this graph before declaring that it shows the opposite of what it actually shows? Was this just an opportunity to harp on a standard news headline - Japanese people aren't having kids, Japan is dying, etc. etc. - with the graph as just an excuse to re-up the news one more time?? Bat Boy is not pleased! Ana Swanson, our policy is to forgive all crazy blog posts (we all do them from time to time), but you must repent and say fifty Hail BatBoys. Conrad Hackett and everyone else who tweeted it without looking, you get a Bat Boy runner-up ribbon, and you are required to say twenty Our BatBoys.

More importantly, when are people going to start writing news articles pointing out that Japan's fertility is on the rise?

Maybe I better do it.


  1. Anonymous2:08 AM

    For most people forecasting is taking a ruler a extending the trendline out to infinity.

    1. Anonymous12:24 PM

      A ruler?! What am I, a scientist? Extending the line freehand was good enough for our forefathers and it's good enough for me.

  2. Yes the fertility rate is up from low of 1.2 in 2010 to 1.39 in 2012 but as the population ages the number of women in the child bearing age bracket is reducing therefore the actual number of births continues to decline. The change in fertility rate will have a minimal impact on the demographic issues facing Japan.

    1. HK...

      On the mark! Noah is confusing rates with actual population growth/decline. But, he has to be forgiven as he is infatuated with Japan. I am waiting for Noah to renounce his citizenship and move to Japan where he could engage in improving fertility rate.

    2. Anonymous6:13 PM

      Fertility changes precede demographic changes.

  3. Anonymous2:27 AM

    I'm waiting for someone to point out that the term, 'fertility rate' is just a horrible description for what is measured. Fertility has had nothing to do with the birth rate for a long, long time. This isn't a question of fertility.

    Why is this important? Because the Japanese press makes it sound like nobody in Japan is having sex, which I don't think is true. Not that it matters how often people have sex these days in relation to how many children are born, but even so, this whole subject is hexed by the poor name.

    These days most people choose quite specifically when they will have children, and the frequency of sex is not related to that. I'm not bragging, but I've chosen to have 2 children and each time it took 1 month max to produce the desired outcome through the abstinence of birth control.

    Heck, I'd have 15+ children if it weren't for birth control. Fertility? Not a problem.

    Time? A significant problem!

    1. "Because the Japanese press makes it sound like nobody in Japan is having sex, which I don't think is true."

      Well, a few are...but not many. See this Mark Steyn article on this very 'problem':

    2. You said: "Because the Japanese press makes it sound like nobody in Japan is having sex, which I don't think is true. Not that it matters how often people have sex these days in relation to how many children are born, but even so, this whole subject is hexed by the poor name."

      Well, there are two points:

      1) No, the Japanese aren't having sex

      More than 40 per cent of Japan's adult singles are virgins, says study

      2) Yes, it matters. They are, apparently, uninterested.

  4. As noted, rising from 1.2 in 2005 up to almost 1.4 in 2012/ 2014 will not much reduce Japan's demographic neutron bomb. What is the projected number of people in the 0-19; 20-39; 40-59; 60-79; 80+ "quintiles" of dependents (1, 5, & some 2 & 4) vs prime workers?

    On the other hand, if robots do almost all the work in Japan, especially making great cars and other exportables, the dependents might all have some kind of "jobs" and a very low rate of real poverty, tho an increasing number of old folks will wanting more health care (care, not insurance).

  5. Anonymous5:40 AM

    But what you point out still adds up to : ZERO !!

    Abuse of words has been the great instrument of sophistry and chicanery, of party, faction, and division of society.
    john adams

  6. Anonymous6:00 AM

    "More importantly, when are people going to start writing news articles pointing out that Japan's fertility is on the rise?"

    Aging is part of a big research agenda that really started at the OECD. It is also linked to overstated worries about pension and budget deficits. Once these people have retired from their tenured positions, MAYBE, we can get a sensible discussion. Unless we get wholesale sackings at places like MIT, the Minnesota Fed and the OECD, we are not going to get relevant economics back.

  7. Anonymous9:39 AM

    I remember seeing a paper a few years back that pointed out this was happening across all major developed countries. I can't seem to find it now, though, but here's one that makes a similar point:

    1. Thanks! Yeah, I've seen that too I think, but I wondered whether in Europe it was driven mainly by immigrants...

  8. What is the lower bound on fertility rates for a prosperous and industrialized nation? Do we have any precedent for such long periods of declining fertility for a nation already below replacement rate?

  9. Anonymous12:35 PM

    Despite this point, the excessive exuberance about the rising Japan of the 1980's led to some real chestnuts regarding Japanese culture and society, including the birth rate. There were some incredibly optimistic views on birthrates back then, and even more excessive optimism by some regarding Japan's supposed immunity to the evils of modernity...which should have been obvious bullocks by 1990, but wasn't.

    Takeshi Umehara, usually quite good, made some very over optimistic statements in the early 90's regarding Japanese culture in his piece "The Civilization of the Forest: Ancient Japan Shows Postmodernism the Way". He basically said that Japan was just as premodern as an Amazonian tribe, but was somehow more technologically advanced at the same time.

    Also, for a less sophisticated perspective, read some of the comments on young women in the book "Gentle Ways in Japan"...a very quasi orientalist photo book, which manages to imply that because all these young women said they wanted to be mothers, they would actually *be mothers* after they graduate high school.

    Anyways, I should read the piece, but I think there is a real lesson to be taken from this exuberance: don't overestimate any trend, and don't succumb to wishful thinking.

  10. From the chart, it appears that about every 20 years or so, fertility rate seems to jump and then peters out. Take a look at/around 1965, 1985 and 2005. Any explanation on thsi?

    1. Anonymous8:44 PM

      I bet if you and grey "enlightenment" put your heads together you could generate some pretty mind-blowing derp.

    2. I wish you would comment without being Anonymous - just say A1 or An based on the flavor available.

      My point was that one can make a chart to say whatever... But, come to think of it, Japanese kids may be repressed until they all get to be adult and behave wildly, petering away their youth, like every generation or so?

  11. from wikipedia

    Herbivore men (草食(系)男子 Sōshoku(-kei) danshi?) is a social phenomenon in Japan of men who shun marriage or gaining a girlfriend.[8] This phenomenon is viewed by the Japanese government as a leading cause in the nation's declining birth rate, prompting the government to provide incentives for couples that have children, including payouts and free health care.[9]

  12. Anonymous6:15 PM

    A couple of decades ago it was overpopulation. Now it's underpopulation.

  13. I don't have it offhand, but every few years there's a survey on intentions for the number of children (it may only be published in Japanese). That's barely budged from 2.0 over the last 50 years.

    What has shifted is female LF participation, which went from 43% for age 25-29 in 1975 to 79% in 2013, while for age 30-34 that ratio's gone from 44% to 70%. Marriage is delayed, from average of age 24 to age 29, and childbearing is delayed once married. In 2010 peak fertility was at age 30 at 106 births per 1,000 women, in 1960 the peak was at age 26 at 202 births per 1,000 women. Over the past decade fertility has declined for women under age 32. So the reality is that despite their stated intention the combination of late marriage and a delay in the start of childbearing within marriage means the number of women who hit their target is low.

    Note TFR is a synthetic measure, summing age-specific fertility at a point in time, and not weighting that for the numbers of women of a specific age. In effect, it assumes that women entering their child-bearing years, e.g. 15-year-olds in 2010, will have the same fertility rate at age 30 [that is, in 2025] as 30-year-old women have in 2010.

    Two additional minor points. Childbearing out of wedlock remains very low (though in 40% of marriages the first child is born less than 10 months later -- why bother with the legal formality until a child is involved?). Second, 50 years ago marriage was almost universal; the number of women never married was under 2% in 1960, is 11% today.

    All these data are from one of the periodic population handbooks [the 人口統計資料集2013], eg Table T04-06 for age-specific fertility from 1930 to 2010, for every 5th year from 1950).

  14. Japanese National Fertility Survey, every 5 years. In the 7th survey (1977) women age 20-24 wanted 2.19 children, those age 30-34 wanted 2.21. Reversed today: in the 14th survey (2010) women age 20-24 wanted 2.45 children, but age 30-34 want 2.15. No idea what the confidence interval is, but a constant "2 children" across age brackets and across 25 years is my sense of a safe reading.

  15. Hey, who are you calling "you all"